WO2010091640A1 - 一种多端口负载分担方法、装置和网络系统 - Google Patents

一种多端口负载分担方法、装置和网络系统 Download PDF

Info

Publication number
WO2010091640A1
WO2010091640A1 PCT/CN2010/070632 CN2010070632W WO2010091640A1 WO 2010091640 A1 WO2010091640 A1 WO 2010091640A1 CN 2010070632 W CN2010070632 W CN 2010070632W WO 2010091640 A1 WO2010091640 A1 WO 2010091640A1
Authority
WO
WIPO (PCT)
Prior art keywords
port
physical
load sharing
logical
sub
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2010/070632
Other languages
English (en)
French (fr)
Inventor
颜清华
柏璐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP10740929.4A priority Critical patent/EP2391085B1/en
Publication of WO2010091640A1 publication Critical patent/WO2010091640A1/zh
Priority to US13/209,043 priority patent/US8848717B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/24—Multipath
    • H04L45/245—Link aggregation, e.g. trunking
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/28—Routing or path finding of packets in data switching networks using route fault recovery
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/12—Avoiding congestion; Recovering from congestion
    • H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/14—Multichannel or multilink protocols
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q11/00—Selecting arrangements for multiplex systems
    • H04Q11/0001—Selecting arrangements for multiplex systems using optical switching
    • H04Q11/0062—Network aspects
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q11/00—Selecting arrangements for multiplex systems
    • H04Q11/0001—Selecting arrangements for multiplex systems using optical switching
    • H04Q11/0062—Network aspects
    • H04Q2011/0079—Operation or maintenance aspects
    • H04Q2011/0081—Fault tolerance; Redundancy; Recovery; Reconfigurability
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Definitions

  • a multi-port load sharing method, device and network system The present application claims a Chinese patent application filed on February 13, 2009, with the application number 200910009026.1, the invention is a multi-port load sharing method, device and network system. Preferably, the entire contents of which are incorporated herein by reference.
  • the present invention relates to network communication technologies, and in particular, to a multi-port load sharing method, apparatus, and network system.
  • the optical device In the network structure of the current telecommunication carrier, the optical device is used to form a physical transmission network, and the data device is directly connected to the optical device as a user layer device. The connection between the data devices is actually achieved through the physical channel of the optical device.
  • the interface between the data device (such as a router) and the network device (such as an optical device) is the U I interface (User Network Interface). The link between the two is the U I link.
  • Commonly used U I interface types include Ethernet interface, POS interface (Packet over SDH/SONET, mapping interface for transmitting data packets on SDH/SONET), and OTN (Optical Transport Network) interface.
  • SDH/SONET Synchronous Digital Hierarchy I Synchronous Optical Network Synchronous
  • Common Ethernet interface rates are 1Gbps, 10Gbps, and the standard organization is defining 100Gbps, etc.
  • POS interface rates are 155Mbps, 622Mbps, 2.5Gbps, 10OGbps and 40Gbps
  • OTN interface rates are 2.5Gbps, 10OGbps, 40Gbps, and standard organization The lOOGbps being defined, etc.
  • the main function of the data device is to forward the data packets from the link layer or the network layer to the appropriate port according to the routing algorithm.
  • the optical device configures the cross-connection on the physical channel layer according to the service requirement, and selects the corresponding port to connect to the optical fiber through the optical port. Other optical equipment.
  • Link Aggregation is proposed by the standard IEEE 802.3ad to combine two or more data channels into a single channel with a single higher bandwidth logical link. Appeared. Link Aggregation is generally used to meet the needs of large bandwidth connections, and the protection mechanism of intra-group ports is implemented through redundant physical ports. The main idea of this technology is to assign the MAC addresses of all physical ports in the group to the same network layer port, that is, multiple ports have only one IP address, and only one destination IP address.
  • the physical port is configured as a trunk port (Tmnked Port), and the corresponding Tmnked Port is also connected to the peer end to form an aggregated link (Aggregated Link).
  • the physical ports D1-P1 and D1-P2 of the data device 101 are configured as Tmnked Port; the physical ports D2-P1 and D2-P2 of the peer data device 102 are correspondingly configured as Tmnked Port, and the data device 101 A physical link is formed between the physical port D1-P1 and the physical port D2-P1 of the peer data device 102, and another physical port D1-P2 of the data device 101 and the physical port D2-P2 of the peer data device 102 form another A physical link.
  • the physical link between the Tmnked Ports of the two data devices forms an Aggregated Link. It can be seen that the Aggregated Link is actually composed of two physical links. When one physical link fails, the other physical link can protect.
  • the two physical links have the same source and the same destination, that is, the source IP address and the destination IP address of the two physical links are the same.
  • the disadvantages of the prior art Link Aggregation are mainly: Because the two physical links with the same destination need to be bound together to protect, the UI link with different destinations cannot be protected; single layer network protection, The impact of the failure of the UI link on the entire network cannot be isolated. As the network bandwidth continues to increase and the port capacity continues to increase, the requirements of the network for the reliability of the network, the QOS (Quality Of Service) of the data service, the network expansion, and the initial network construction cost are also increased. The higher it is. The existing technology has been unable to meet the operator's demand for the network.
  • QOS Quality Of Service
  • the embodiments of the present invention provide a multi-port load balancing method, device, and network system, which can solve the protection problem of the U I link connecting different destinations, limit the impact of the U I link failure on the entire network, and avoid network oscillation.
  • An embodiment of the present invention provides a multi-port load balancing method, including: receiving a data packet sent to a destination end; and searching for an address port mapping table according to the address information of the destination end carried by the data packet Corresponding to the logical outbound port information; obtaining the information of the corresponding load sharing group according to the logical outbound port information searching logical port table, where the load sharing group includes at least two physical ports, and each physical port corresponds to at least two subchannels;
  • the preset load sharing policy encapsulates the data packets in at least one sub-channel corresponding to each physical port in the load sharing group and forwards the data to the destination end through the corresponding physical port.
  • An embodiment of the present invention provides a data device for implementing multi-port load balancing, including: a receiving module, configured to receive a data packet sent to a destination end; and an address port mapping table searching module, configured to be carried according to the data packet The address information of the destination is searched for the address port mapping table to obtain the corresponding logical outbound port information.
  • the logical port table searching module is configured to obtain the information of the corresponding load sharing group according to the logical outbound port information searching logical port table, where the load sharing group includes At least two physical ports, each physical port corresponding to at least two sub-channels; and a sending module, configured to encapsulate the data packets in each of the physical ports corresponding to the load sharing group according to a preset load sharing policy A sub-channel is forwarded to the destination through a corresponding physical port.
  • An embodiment of the present invention provides a network system for implementing multi-port load sharing, including a data device and a network device, where the data device is configured to receive a data packet sent to a destination end, according to a destination carried in the data packet.
  • the address information lookup address port mapping table of the terminal obtains the corresponding logical outbound port information, and obtains information about the corresponding load sharing group according to the logical outbound port information search logical port table, where the load sharing group includes at least two physical ports, each The physical port is corresponding to the at least two sub-channels, and the data packets are respectively encapsulated in at least one sub-channel corresponding to each physical port in the load sharing group according to a preset load sharing policy, and sent to the network through the corresponding physical port.
  • the destination is sent directly to the destination.
  • the multi-port load sharing method, device, and network system provided by the embodiment of the present invention share the data packet in at least one sub-channel corresponding to one or more physical ports in the same load sharing group, so that the data packet can pass Different sub-channels reach the same or different network device ports.
  • a UI link fails, data packets can reach the destination address through other sub-channels. Therefore, embodiments of the present invention improve network reliability and achieve different connections. of destination Protection of the UNI link. Since the data packet can be transmitted through multiple sub-channels of the plurality of physical ports, the embodiment of the present invention also solves the problem that the core network has a large capacity port shortage, and the embodiment of the present invention can reduce the initial investment during network construction. , can also achieve smooth expansion.
  • FIG. 2 is a schematic flowchart of a multi-port load sharing method according to an embodiment of the present invention
  • FIG. 3 is a schematic diagram of a network model for implementing a multi-port load sharing method according to an embodiment of the present invention
  • FIG. 4 is a schematic diagram of a recovery route using a multi-port load sharing method according to an embodiment of the present invention
  • FIG. 6 is a schematic diagram of UI link protection using the multi-port load sharing method according to the embodiment of the present invention
  • FIG. 7 is a multi-port load sharing method for implementing multiple network devices according to the embodiment of the present invention
  • FIG. 8 is a schematic diagram of port connection of an application example of a multi-port load sharing method according to an embodiment of the present invention.
  • FIG. 9 is a schematic diagram of port connection of an application example of a multi-port load sharing method according to an embodiment of the present invention.
  • FIG. 10 is a schematic diagram of an application example of interconnecting data devices of different capacity ports by using the multi-port load sharing method of the embodiment of the present invention.
  • FIG. 11 is a schematic diagram of an application example of establishing a tunnel path by using the multi-port load sharing method of the embodiment of the present invention.
  • FIG. 12 is a schematic diagram of an application example of establishing a tunnel path by using the multi-port load sharing method according to the embodiment of the present invention.
  • FIG. 13 is a schematic structural diagram of a data device for implementing multi-port load sharing according to an embodiment of the present invention.
  • FIG. 14 is a schematic structural diagram of a network system for implementing multi-port load sharing according to an embodiment of the present invention.
  • an embodiment of the present invention provides a multi-port load sharing method, including:
  • the address information of the destination end may be an IP address or a MAC address of the destination end; correspondingly, the address port mapping table may be corresponding to a routing table or a MAC table;
  • the data packet is encapsulated in at least one physical sub-channel corresponding to each physical port in the load sharing group according to a preset load sharing policy, and forwarded to the destination end through the corresponding physical port.
  • the load sharing group in the embodiment of the present invention usually includes multiple physical ports, and is therefore often referred to as a multi-port load sharing group.
  • the sub-channel corresponding to the physical port is not limited to a physical sub-channel, and may also be a logical sub-channel or a sub-channel including a physical layer transmission frame structure. And logical subchannels. Since the method in the foregoing embodiment implements load balancing by using at least one physical sub-channel corresponding to different physical ports, the load sharing group can also be understood as a set of physical sub-channels for sharing data packets.
  • the sub-channel includes: an ODU series corresponding to the OTN interface, a VC-4 series corresponding to the POS interface, or a channelized interface corresponding to the 100GE interface.
  • the physical subchannel can be VC-4 in an STM-4 frame.
  • the multiple ports participating in the Payload Plan is referred to as the MP4 technology
  • the multi-port load sharing group is simply referred to as the MP4 group.
  • data packets destined for the same destination can be carried by different physical ports in the MP4 group. Moreover, data packets in each physical subchannel corresponding to the same physical port can be sent to the same or different destinations.
  • the capacity of a logical port may be the same as or different from the capacity of a physical port.
  • the number of logical ports may be equal or unequal to the number of physical ports.
  • the MP4 method in the embodiment of the present invention may be implemented by a data device, such as a router, a switch, or the like.
  • the method in the embodiment of the present invention may be further applied to a network scenario in which data transmission is performed across a transmission network, where the transmission network may be an optical transmission network, and the network device in the transmission network is corresponding to an optical device.
  • the method specifically includes: after receiving the data packet sent by the source end to the destination end, the data device at the source end side searches the address port mapping table according to the address information of the destination end carried by the data packet, and obtains corresponding logical outgoing port information, and according to The logical outbound port information is further searched from the logical port table to obtain a corresponding load sharing group, where the load sharing group includes at least two physical ports, and each physical port corresponds to at least two physical sub-channels;
  • the load balancing policy is configured to encapsulate the data packet in at least one physical sub-channel corresponding to each physical port in the load sharing group, and forward the optical forwarding network to the destination end through the corresponding physical port or directly to the destination.
  • the terminal forwards the data packet.
  • the process of forwarding the data packet directly to the destination end by the optical transmission network includes: when the data device on the source side and the data device on the destination side support the MP4 method in the embodiment of the present invention, the optical transmission network will The data packets sent by different physical ports of the data device on the source side are respectively sent to the physical port corresponding to the data device on the destination end side, and then the data device on the destination end side decapsulates the data packet to obtain the destination.
  • the end-transport network includes multiple optical devices, and data packets sent to the same destination in the MP4 group can be sent to Different physical ports of the same optical device or physical ports of different optical devices.
  • the data packets sent to the same destination end in the MP4 group may be aggregated by any optical device in the optical transmission network, and then forwarded to the data device of the destination end side or directly to the data device of the destination end side.
  • FIG. 3 is a schematic diagram of a network model for implementing a multi-port load sharing method according to an embodiment of the present invention.
  • the data device 101 includes two logical ports and two physical ports.
  • the two logical ports are respectively labeled as D-L1 and D-L2, and the two physical ports are respectively labeled as D-P1 and D-P2, and each physical port corresponds to two physical sub-channels respectively.
  • the physical port D-P1 corresponds to the physical sub-channels SC-1 and SC-2, and the physical port D-P2 corresponds to the physical sub-channels SC-3 and SC-4.
  • the data device 101 is connected to the optical device 102 in the optical transmission network.
  • the physical port D-P1 of the data device 101 is connected to the physical port 0-P1 of the optical device 102 to form a UI link, and another physical port D of the data device 101.
  • -P1 is associated with the physical port 0-P2 of the optical device 102 to form another UNI link.
  • the sum of the load capacities of the physical sub-channels SC-1 and SC-2 can occupy all the bandwidths of the UNI links between D-P1 and 0-P1, and the proportion of the bandwidth occupied by the physical sub-channels SC-1 and SC-2 It can be adjusted according to specific needs, for example, it can be adjusted to be the same or different, and the adjustment can adopt static configuration or dynamic configuration.
  • the physical subchannels SC-3 and SC-4 are similar, and will not be described again.
  • the physical ports D-P1 and D-P2 are configured into one MP4 group.
  • the data packet passing through the logical port D-L1 of the data device 101 will be encapsulated in the physical sub-channels SC-1 and SC-3 and passed through the corresponding physical ports D-P1 and D-, respectively.
  • the P2 is forwarded, and the data packets encapsulated in the physical sub-channels SC-1 and SC-3 are respectively sent to the optical device along the two UI links, and the optical device will be respectively from the corresponding physical port 0-P1 and 0-P2 receives the data packet, and aggregates the data packets encapsulated in the physical sub-channels SC-1 and SC-3 according to the service flow of the data packet, and then sends the data packet from the optical port 0-P4 of the optical device.
  • the data packets passing through the logical port D-L2 of the data device 101 will be encapsulated in the physical sub-channels SC-2 and SC-4 and forwarded through the corresponding physical ports D-P1 and D-P2, respectively.
  • the data packets encapsulated in the physical sub-channels SC-2 and SC-4 are respectively sent to the optical device along two UI links, and the optical device will receive the data from the corresponding physical ports 0-P1 and 0-P2, respectively.
  • the method for acquiring the physical sub-channel for carrying the forwarding data packet at the logical ports D-L1 and D-L2 may use technologies such as existing Trunking or ECMP (Equity Cost Multi-path). Or use other feasible options.
  • the number of physical ports in the MP4 group can be more than two in Figure 3, and the MP4 group can be configured according to the same principle.
  • the sub-channel corresponding to the physical port in the embodiment of the present invention is not limited to a physical sub-channel, and may also be a logical sub-channel or a sub-channel and a logical sub-channel that simultaneously include a physical layer transmission frame structure.
  • the signal mapping format of the physical port may be all interfaces that can be channelized, such as a POS interface or an OTN interface.
  • the hierarchical rate of all POS interfaces or OTN interfaces is applicable to this design.
  • the 100GE interface that the standards organization is developing, using lOxlOGE or 4x25GE, can also implement channelized interfaces.
  • This MP4 scheme can be manually statically configured or dynamically configured by the device.
  • GMPLS General Multi-Protocol Label Switching
  • the same effect can be achieved by manually configuring a cross-connection according to the MP4 scheme. Wait. If the load sharing group includes more than one physical port, the capacity of each physical port may be the same or different.
  • MP4 group configuration schemes According to factors such as physical port capacity, logical port capacity, and load sharing ratio, four major MP4 group configuration schemes can be roughly divided. These MP4 group configuration schemes are described in detail below.
  • D-P1 and D-P2 of the data device 101 and the optical device 102 are both 40G OTN interfaces.
  • D-P1 and D-P2 are configured as MP4 groups containing 2 ports on the data device, and loads from logical ports D-L1 and D-L2 are evenly distributed among the 2 physical ports.
  • the load ODU3 of the 40G OTN interface can be composed of four physical sub-channels of the load ODU2.
  • the load capacity of each physical sub-channel is 10G, which are labeled as #1 0DU2, #2 0DU2, #3 ODU2, and #4 ODU2.
  • Table 1 describes the port load sharing channel relationships of MP4 groups consisting of two 40G OTN interfaces. Two physical ports D-P1 and D-P2 distribute the load equally on the logical ports, each accounting for 50%.
  • Solution 2 The port capacity is the same, and the load is not evenly shared.
  • Table 2 describes the unbalanced load-sharing path relationship of the MP4 group consisting of two 40G OTN interfaces.
  • the load sharing of the logical ports on the physical ports D-P1 and D-P2 is implemented in a 3:1 ratio.
  • Table 1 and Table 2 describe the load distribution schemes with the same port capacity in the MP4 group. In fact, physical ports of different capacities can also be configured as MP4 groups.
  • Table 3 lists the load distribution schemes for three 40G OTN interfaces and one 100G OTN interface. The 100G OTN interface is assumed to be channelized into 10 10G, of which optical equipment also has three 40G ports and one 100G port.
  • Table 4 lists the load distribution scheme for a data device with four 10G logical ports and two 40G physical ports.
  • the optical device has two 40G physical ports.
  • D-L4 M 0DU2 10G For other OTN interfaces, for example, 10G, 100G, etc., you can configure the corresponding physical sub-channels in the same way. Similarly, for channelized POS interfaces, you can also configure the corresponding principles, such as 40G POS interfaces. That is, STM-256/OC-768 level, load sharing can be performed in the physical sub-channel at a rate of 10G, that is, the capacity of VC-4-64c. In the lower-rate POS interface, the load balancing scheme can be configured through the physical sub-channel VC-4.
  • the load sharing configuration may be implemented according to the method in the embodiment of the present invention, as long as the physical sub-channel division can be performed.
  • a physical port group usually includes 2, 4, 8, or 16 ports, etc.
  • the number of physical ports in the MP4 group may be any number greater than or equal to 1.
  • selecting ports in different cards to form an MP4 group can further improve link reliability.
  • RA, RB, RZ, and RX are data devices
  • OA, OB, OC, OD, OE, OZ, OX, OA1, and OA2 are optical devices connected to corresponding data devices, respectively.
  • Al, A2, Bl, B2, and Zl, Z2 are UNI links corresponding to different locations. There is a network optical link interconnection between each optical device.
  • This application example is a specific application for enhancing network reliability by the multi-port load sharing method of the embodiment of the present invention.
  • FIG. 4 it is a schematic diagram of a recovery route that does not use the multi-port load sharing method of the embodiment of the present invention.
  • the route from the data device RA to the data device RB is RA->A1 -> OA -> OB -> B 1 -> RB.
  • the routing of the data device RA to the data device RB is interrupted.
  • the alternate route is RA -> ⁇ 2-> ⁇ -> ⁇ ->Z2->RZ-> Zl -> ⁇ -> ⁇ ->B2->RB.
  • FIG. 5 a recovery route diagram of the multi-port load sharing method according to the embodiment of the present invention is used.
  • the routing from the data device RA to the data device RB includes RA->A1 ->OA->OB->B 1 -> RB and RA->A2->OA->OB->Bl->RB.
  • the UNI link A1 fails, the RA to RB route is not interrupted.
  • the data device RA will update the load sharing group information in the local logical port table according to the failure.
  • the RA When the RA receives the data packet sent to the RB, the RA searches the address port mapping table according to the address information of the RB carried by the data packet to obtain the corresponding logical outgoing port information, and then searches the logical port table according to the logical outgoing port information.
  • the load sharing group information in which the load sharing group only includes the physical port corresponding to the UI link A2, so the data device RA will encapsulate the data packet in at least one physical body corresponding to the physical port according to a preset load sharing policy.
  • the sub-channel is forwarded to the optical device OA through the physical port and the UI link A2, and the data packet is forwarded to the RB by the OA along the route of OA->OB->Bl->RB.
  • the network of the MP4 technology can isolate the propagation of the fault on the network, limit the impact of the UI link failure to a single node, and maintain the original on the network. Routing, thereby saving network resources and avoiding network fluctuations.
  • the solution used in this embodiment fully utilizes the interaction advantages of the data device and the optical device, and can achieve the protection purpose without device or link switching.
  • This application example enhances the protection of the UNI link by the multi-port load sharing method of the embodiment of the present invention.
  • the link D-P2 to 0-P2 fails, and the fault may be a D-P2 port failure.
  • the physical sub-channels SC-3 and SC-4 corresponding to the link fail, the data packets transmitted through the physical ports 0-P3 of the optical device can still be encapsulated in the physical sub-channel SC-2 via the physics of the data device 101.
  • Port D-P1 is forwarded to the physical port 0-P1 of the optical device, and then transmitted to the physical port 0-P3 by the physical port 0-P1 of the optical device.
  • the data packet transmitted through the physical port 0-P4 of the optical device can still be
  • the encapsulation in the physical sub-channel SC-1 is forwarded to the physical port 0-P1 of the optical device via the physical port D-P1 of the data device 101, and then transmitted to the physical port 0-P4 by the physical port 0-P1 of the optical device.
  • the data device and the optical device do not need to perform any switching operation to achieve the protection purpose, and no additional protection resources are needed.
  • FIG. 6 is a schematic diagram of U I link protection according to the multi-port load sharing method of the embodiment of the present invention.
  • Figure 6 shows the difference between MP4 protection and Tmnking protection.
  • the UI link of the MP4 group can protect all UI links between the data device and the optical device without restriction; Tmnking protection has significant limitations and can only protect UNI links to the same destination address.
  • each port in the group is protected by each other.
  • the protected U I link can go to different destinations.
  • the multi-port load sharing method in the embodiment of the present invention can implement adaptive adjustment of the load carried by the physical sub-channel corresponding to different physical ports in the same MP4 group.
  • This application example is to implement multiple network design by using the multi-port load sharing method of the embodiment of the present invention. Prepared UNI link protection.
  • the data device router is taken as an example.
  • the router RA has two physical ports, labeled D-P1 and D-P2, which are connected to different optical devices OA1 and OA2, respectively.
  • the RA needs to establish two links with the router RB and RC.
  • the physical ports D-P1 and D-P2 can also be configured as MP4 port groups.
  • the router RA can forward the data packet to the RB through two routes labeled S1 and S2, where route S1 is RA->OAl->OB->RB, and route S2 is RA->OA2->OB->RB.
  • the RA encapsulates the data packets forwarded to the RB in at least one physical subchannel corresponding to the physical ports D-P1 and D-P2 according to a preset load sharing policy, and then encapsulates the data packets in the physical port D-P1.
  • the data packet in the corresponding physical sub-channel is forwarded to the optical device OA1 through the physical port D-P1
  • the data packet encapsulated in the physical sub-channel corresponding to the physical port D-P2 is forwarded to the optical device through the physical port D-P2.
  • the optical device OB aggregates the data packets from the two different routes and then forwards the data packets to the router RB through the same optical device physical port. It can be seen that even if the UNI link between the router RA and the optical device OA1 fails, the route S1 is interrupted, and the data packet forwarded by the router RA to the router RB can still be forwarded by using the route S2, that is, from the router RA to the optical device.
  • the OA2 is sent to the router RB through the optical device OA2 to achieve the protection purpose.
  • the data link protection mechanism of the router RA to the router RC is the same as the data link protection mechanism of the router RA to the router RB.
  • the data packet can be transmitted through the load sharing paths S3 and S4, and will not be described here.
  • the multi-port load sharing method of the embodiment of the present invention when the multi-port load sharing method of the embodiment of the present invention is applied to U I link protection of multiple network devices, it has stronger protection capability. Because this application not only protects the link failure of the U I, but also protects the node failure of the network device. As shown in Figure 7 for optical device OA1 or OA2, no matter which optical device fails, this fault can be protected either by a port failure or by the device itself. In the actual network application, the multi-port load sharing method in the embodiment of the present invention can also be applied to U I link protection between multiple data devices and multiple optical devices.
  • Application example four This application example is a specific application for saving the number of physical ports by the multi-port load sharing method of the embodiment of the present invention.
  • the mesh network structure is usually adopted. All the nodes in the network are interconnected, and a large capacity port is required, and the actual proportion of the total bandwidth occupied by the service is not high.
  • an optical device connected to a data device not only does the client side require many ports, but also a large number of physical ports on the same line side, so it is especially important to save the number of ports.
  • Figure 8 is a schematic diagram of port connections without the MP4 scheme
  • Figure 9 is a schematic diagram of port connections using the MP4 scheme. Comparing Figure 8 with Figure 9, it can be seen that after using the MP4 scheme, both the data device and the optical device can save a large number of physical ports.
  • Figure 8 and Figure 9 are mesh networks with five nodes. Before using the MP4 solution, each router needs four physical ports to connect with optical devices. After using the MP4 solution, take RA as an example, as shown in Figure 9.
  • the RA is configured as a load sharing group by connecting a physical port that is connected to the OA to form the UNI link A1 and a physical port that is connected to the OA to form the UI link A2, so that the RA is sent to the RB, RC, RD, and RE, respectively.
  • the data packet of any of the data devices may be encapsulated in at least one physical sub-channel corresponding to the UI links A1 and A2, and sent to the optical device OA through the UI links A1 and A2, and then forwarded to the corresponding OA by the OA.
  • the router can save up to ten ports; the optical device also saves the same number of ports. As the number of network nodes increases, the number of physical ports saved will be more.
  • the multi-port load sharing method of the embodiment of the present invention can also achieve the purpose of reducing initial investment and smoothing capacity expansion.
  • This application example is a specific application in which the multi-port load sharing method of the embodiment of the present invention does not directly relate the interconnect bandwidth of the data device to the port capacity.
  • the router RA has a physical interface with a 10G rate.
  • the router RB has a physical interface of 40G.
  • the RA and the RB can have 40G bandwidth interconnection. The RB no longer needs to configure a service card with a 10G rate.
  • the logical interconnection between data devices and the interconnection of data devices and optical devices require the same physical port capacity to achieve a normal connection.
  • the logical interconnect bandwidth between data devices can be connected in a variety of flexible ways, even through the convergence of physical sub-channels to achieve low-end products (no high-capacity ports) and high-end products connected at high rates.
  • the application example is a specific manner of implementing a tunneling scheme superior to TE (Traffic Engineering) by using the multi-port load sharing method of the embodiment of the present invention.
  • Figures 11 and 12 show the comparison before and after the MP4 scheme.
  • the RA-RZ ramp needs to be established.
  • the path S1 in Figure 11 represents the actual physical path through which the ramp is established.
  • S2 and S3 represent the physical path shared by the tunnel but the single-hop of the different sub-channels. routing.
  • the established ramp can be directly implemented through the physical sub-channel S4 of the optical device, and does not require the participation of the intermediate router, thereby saving a large amount of data equipment resources.
  • an embodiment of the present invention further provides a data device for implementing multi-port load sharing, including:
  • the receiving module 131 is configured to receive a data packet sent to the destination end;
  • the address port mapping table searching module 132 is configured to obtain the corresponding logical outgoing port information by searching the address port mapping table according to the address information of the destination end carried by the data packet;
  • the logical port table searching module 133 is configured to obtain the information about the corresponding load sharing group according to the logical out port information searching logical port table, where the load sharing group includes at least two physical ports, and each physical port corresponds to at least two physical ports. Subchannel
  • the sending module 134 is configured to encapsulate the data packet into at least one physical sub-channel corresponding to each physical port in the load sharing group according to a preset load sharing policy, and forward the data packet to the destination end through the corresponding physical port.
  • the sub-channel corresponding to the physical port in the embodiment of the present invention is not limited to the physical sub-channel.
  • the sub-channel includes an ODU series corresponding to the OTN interface, a VC-4 series corresponding to the POS interface, or a channelized interface corresponding to the 100GE interface.
  • the data device provided by the embodiment of the present invention may be a router or a switch.
  • the data device can implement multi-port load sharing by referring to the method shown in FIG. 2 in the method embodiment of the present invention.
  • an embodiment of the present invention further provides a network system for implementing multi-port load sharing, including a data device 141 and a network device 142, where:
  • the data device 141 is configured to receive a data packet sent to the destination end, obtain an address corresponding to the logical port according to the address port mapping table of the destination information carried in the data packet, and obtain a logic according to the logical out port information.
  • the port table obtains information about the corresponding load sharing group, where the load sharing group includes at least two physical ports, and each physical port corresponds to at least two physical sub-ports.
  • the channel is encapsulated in the at least one physical sub-channel corresponding to each physical port in the load sharing group according to the preset load sharing policy, and is forwarded to the network device 142 through the corresponding physical port; Or send it directly to the destination.
  • the sub-channel corresponding to the physical port in the embodiment of the present invention is not limited to the physical sub-channel.
  • the network device 142 is an optical device, and the data device may be a router, a switch, or the like.
  • the network system of the embodiment of the present invention can implement the multi-port load balancing by referring to the method shown in FIG. 2 in the method embodiment of the present invention.
  • the networking modes shown in the foregoing application example 1 to application example 6 can be used to form various network forms.
  • the data device and the network system provided by the embodiments of the present invention distribute the load to multiple physical sub-channels corresponding to one or more physical ports in the same MP4 group, so that the load can reach the same through different sub-channels.
  • the network device port when the UI link fails, the data packet can reach the destination address through other sub-channels, so the embodiment of the present invention improves the reliability of the network and realizes the protection of the UI link connecting different destinations. Since the data packet can be transmitted through multiple physical sub-channels of the plurality of physical ports, the embodiment of the present invention also solves the problem that the core network has a large capacity port, and the embodiment of the present invention can be reduced during network construction. Initial investment can also achieve smooth expansion.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明的实施例公开了一种多端口负载分担方法、装置和网络系统，涉及网络通信技术。其中所述方法包括：接收发送给目的端的数据包；根据所述数据包携带的目的端的地址信息从地址端口映射表中查找获得对应逻辑出端口信息；根 据所述逻辑出端口信息从逻辑端口表中查找获取对应负载分担组的信息，其中所述负载分担组包括至少两个物理端口，每个物理端口对应至少两个子通道；根据预设的负载分担策略将所述数据包分别封装于所述负载分担组内每个物理端口所对应的至少一个子通道中并通过对应的物理端口向目的端转发。本发明的实施例适用于网络通信。

Description

一种多端口负载分担方法、 装置和网络系统 本申请要求了 2009年 2月 13 日提交的、 申请号为 200910009026.1、 发 明名称为一种多端口负载分担方法、 装置和网络系统的中国专利申请的优选 权， 其全部内容通过引用结合在本申请中。
技术领域
本发明涉及网络通信技术， 尤其涉及一种多端口负载分担方法、 装置和 网络系统。
背景技术
在当前电信运营商的网络结构中， 光设备用来组建物理传输网络， 数据 设备作为用户层设备与光设备直接相连。 数据设备之间的连接实际上是通过 光设备的物理通道来实现的。 数据设备（如路由器）与网络设备（如光设备） 之间相连的接口即为 U I接口 （ User Network Interface 用户网络接口）， 二者 之间的链路即为 U I链路。
常用的 U I 接口类型， 有以太网接口、 POS 接口 （ Packet over SDH/SONET , 在 SDH/SONET上传送数据包的映射接口）和 OTN ( Optical Transport Network, 光传送网）接口等。其中， SDH/SONET( Synchronous Digital Hierarchy I Synchronous Optical Network Synchronous )是指同步光网络。 常用 的以太网接口速率为 lGbps、 lOGbps, 以及标准组织正在定义的 lOOGbps等； POS接口速率有 155Mbps、 622Mbps, 2.5Gbps、 lOGbps和 40Gbps等； OTN 接口速率有 2.5Gbps、 lOGbps, 40Gbps, 以及标准组织正在定义的 lOOGbps 等。
数据设备的主要功能是将来自链路层或者网络层的数据包根据路由算法 选择合适的端口转发； 光设备则是根据业务需求在物理通道层上配置交叉连 接， 选择相应的端口通过光纤连接至其它光设备。
链路聚合 ( Link Aggregation )是标准 IEEE 802.3ad提出的， 将两个或更 多数据信道结合成一个单个的信道,该信道以一个单个的更高带宽的逻辑链路 出现。 Link Aggregation—般用来满足大带宽连接的需求， 通过冗余的物理端 口来实现组内端口的保护机制。 该技术的主要思想是， 将组内所有物理端口 的 MAC地址分配到同一网络层端口， 即多个端口只有一个 IP地址， 它们的 目的 IP地址也只有一个。 这些物理端口配置成干路端口 （ Tmnked Port ), 对 端也有对应的 Tmnked Port与之连接， 从而形成聚合链路 ( Aggregated Link )。
如图 1所示， 数据设备 101的物理端口 D1-P1和 D1-P2配置成 Tmnked Port; 对端数据设备 102的物理端口 D2-P1和 D2-P2相应地配置成 Tmnked Port,数据设备 101的物理端口 D1-P1和对端数据设备 102的物理端口 D2-P1 之间形成一条物理链路，数据设备 101的物理端口 D1-P2和对端数据设备 102 的物理端口 D2-P2之间形成另一条物理链路。这两台数据设备的 Tmnked Port 之间的物理链路形成 Aggregated Link。 可见， Aggregated Link实际上是由两 条物理链路组成的， 当其中一条物理链路发生故障时， 另一条物理链路可以 起到保护作用。 这两条物理链路有相同的源和相同的目的地， 即这两条物理 链路的源 IP地址和目的 IP地址相同。
现有技术 Link Aggregation的缺点主要是： 由于需要将目的地相同的两条 物理链路绑定在一起才能起到保护作用， 因此对目的地不同的 U I链路不能 提供保护； 单层网络保护， 隔离不了 U I链路的故障对全网的影响。 并且随 着网络带宽的不断增长， 端口容量的不断提升， 运营商对网络的可靠性、 数 据业务的 QOS ( Quality Of Service, 服务质量）、 网络扩容、 以及对初期建网 成本等的要求也越来越高。 现有技术已无法达到运营商对网络的需求。
发明内容
本发明的实施例提供一种多端口负载分担方法、 装置和网络系统， 能够 解决连接不同目的地的 U I链路的保护问题、 限制 U I链路故障对全网造成 的影响且避免网络震荡。
本发明的实施例釆用如下技术方案：
本发明的一实施例一种多端口负载分担方法， 包括： 接收发送给目的端 的数据包； 根据所述数据包携带的目的端的地址信息查找地址端口映射表获 得对应逻辑出端口信息； 根据所述逻辑出端口信息查找逻辑端口表获取对应 负载分担组的信息， 其中所述负载分担组包括至少两个物理端口， 每个物理 端口对应至少两个子通道； 根据预设的负载分担策略将所述数据包分别封装 于所述负载分担组内每个物理端口所对应的至少一个子通道中并通过对应的 物理端口向目的端转发。
本发明的一实施例一种用于实现多端口负载分担的数据设备， 包括： 接 收模块， 用于接收发送给目的端的数据包； 地址端口映射表查找模块， 用于 根据所述数据包携带的目的端的地址信息查找地址端口映射表获得对应逻辑 出端口信息； 逻辑端口表查找模块， 用于根据所述逻辑出端口信息查找逻辑 端口表获取对应负载分担组的信息， 其中所述负载分担组包括至少两个物理 端口， 每个物理端口对应至少两个子通道； 以及发送模块， 用于根据预设负 载分担策略将所述数据包分别封装于所述负载分担组内每个物理端口所对应 的至少一个子通道中并通过对应的物理端口向目的端转发。
本发明的一实施例一种用于实现多端口负载分担的网络系统， 包括数据 设备和网络设备， 其中所述数据设备用于接收发送给目的端的数据包， 根据 所述数据包中携带的目的端的地址信息查找地址端口映射表获得对应逻辑出 端口信息， 并根据所述逻辑出端口信息查找逻辑端口表获取对应负载分担组 的信息， 其中所述负载分担组包括至少两个物理端口， 每个物理端口对应至 少两个子通道， 根据预设负载分担策略将所述数据包分别封装于所述负载分 担组内每个物理端口所对应的至少一个子通道中并通过对应的物理端口向网 发送给目的端或直接发送给目的端。
本发明实施例提供的多端口负载分担方法、 装置和网络系统， 在同一个 负载分担组内， 将数据包通过一个以上的物理端口所分别对应的至少一个子 通道进行分担， 因而数据包能够通过不同的子通道到达相同或不同的网络设 备端口， 在一条 U I链路发生故障时， 数据包能够通过其它的子通道到达目 的地址， 因而本发明的实施例提高了网络的可靠性， 实现连接不同目的地的 UNI链路的保护。 由于数据包可以通过多个物理端口中的多个子通道进行传 输， 因而本发明的实施例还解决了核心网大容量端口不足的问题， 在建网时 釆用本发明的实施例能够减少初期投资， 还能够实现平滑扩容。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案， 下面将对实 施例或现有技术描述中所需要使用的附图作简单地介绍， 显而易见地， 下面 描述中的附图仅仅是本发明的一些实施例， 对于本领域普通技术人员来讲， 在不付出创造性劳动的前提下， 还可以根据这些附图获得其它的附图。
图 1为现有技术链路聚合的示意图；
图 2为本发明的实施例多端口负载分担方法的流程示意图；
图 3为实现本发明实施例多端口负载分担方法的网络模型示意图； 图 4为未釆用本发明实施例多端口负载分担方法的恢复路由示意图； 图 5为釆用本发明实施例多端口负载分担方法的恢复路由示意图； 图 6为釆用本发明实施例的多端口负载分担方法的 U I链路保护示意图； 图 7 为釆用本发明实施例的多端口负载分担方法实现多台网络设备的
U I链路保护的应用实例示意图；
图 8 为未釆用本发明实施例的多端口负载分担方法的应用实例的端口连 接示意图；
图 9 为釆用本发明实施例的多端口负载分担方法的应用实例的端口连接 示意图；
图 10为釆用本发明实施例的多端口负载分担方法实现不同容量端口的数 据设备互连的应用实例示意图；
图 11为未釆用本发明实施例的多端口负载分担方法建立隧道通路的应用 实例示意图；
图 12为釆用本发明实施例的多端口负载分担方法建立隧道通路的应用实 例示意图；
图 13 为本发明实施例用于实现多端口负载分担的数据设备的结构示意 图；
图 14 为本发明实施例用于实现多端口负载分担的网络系统的结构示意 图。
具体实施方式
下面结合附图对本发明实施例一种多端口负载分担方法、 装置和网络系 统进行详细描述。
应当明确， 所描述的实施例仅仅是本发明一部分实施例， 而不是全部的 实施例。 基于本发明中的实施例， 本领域普通技术人员在没有作出创造性劳 动前提下所获得的所有其他实施例， 都属于本发明保护的范围。
如图 2所示， 本发明的实施例提供一种多端口负载分担方法， 包括：
201、 接收发送给目的端的数据包；
202、 根据所述数据包携带的目的端的地址信息查找地址端口映射表获得 对应逻辑出端口信息。
其中所述目的端的地址信息可以为目的端的 IP地址或 MAC地址； 对应 地， 所述地址端口映射表可以对应为路由表或 MAC表；
203、 根据所述逻辑出端口信息查找逻辑端口表获取对应负载分担组的信 息， 其中所述负载分担组包括至少两个物理端口， 每个物理端口对应至少两 个物理子通道；
204、 根据预设的负载分担策略将所述数据包分别封装于所述负载分担组 内每个物理端口所对应的至少一个物理子通道中并通过对应的物理端口向目 的端转发。
本发明实施例中所述负载分担组通常包含多个物理端口， 因此常称为多 端口负载分担组。 可选地， 本发明实施例所述方法在实现负载分担时， 所述 物理端口所对应的子通道并不限于物理子通道， 也可以为逻辑子通道或者同 时包括物理层传输帧结构的子通道和逻辑子通道。 由于上述实施例所述方法 是通过对应不同物理端口的至少一个物理子通道来实现负载分担， 因此所述 负载分担组实际上也可以理解为用于分担数据包的物理子通道的集合。 具体 地， 所述子通道包括： OTN接口对应的 ODU系列， POS接口对应的 VC-4系 列， 或 100GE接口对应的通道化接口。 例如： 物理子通道可以为 STM-4帧中 的 VC-4。
为了描述的方便，以下将多端口负载分担技术（ Multiple ports Participation ofPayload Plan ) 简称为 MP4技术， 将多端口负载分担组简称为 MP4组。
在上述实施例的基础上， 发往同一目的端的数据包可以由 MP4组内的不 同物理端口承担。 并且， 同一物理端口对应的各个物理子通道中的数据包可 以被发送至相同或者不同目的端。 逻辑端口的容量与物理端口的容量可以相 同或者不同， 逻辑端口的数目与物理端口的数目可以相等或不相等。
本发明实施例所述 MP4方法可以由数据设备来实现， 如路由器、 交换机 等。 此外， 本发明实施例所述方法还可以进一步应用于跨越传输网络进行数 据传输的网络场景中， 所述传输网络可以为光传输网络， 所述传输网络中的 网络设备对应为光设备。 所述方法具体包括： 源端侧的数据设备在接收到源 端发送至目的端的数据包后， 根据所述数据包携带的目的端的地址信息查找 地址端口映射表获得对应逻辑出端口信息， 并根据所述逻辑出端口信息从逻 辑端口表中进一步查找获取对应负载分担组， 其中所述负载分担组包括至少 两个物理端口， 每个物理端口对应至少两个物理子通道； 然后， 数据设备根 据预设的负载分担策略将所述数据包分别封装于所述负载分担组内每个物理 端口所对应的至少一个物理子通道中并通过对应的物理端口向光传输网络转 述目的端转发或者直接向目的端转发所述数据包。 所述光传输网络直接向目 的端转发所述数据包的过程具体包括： 在源端侧数据设备和目的端侧数据设 备都支持本发明实施例所述 MP4方法时， 所述光传输网络将从源端侧的数据 设备不同物理端口发来的所述数据包分别发送至目的端侧的数据设备对应的 物理端口， 再由所述目的端侧的数据设备对所述数据包解封装后获得目的端 输网络中包括多个光设备， MP4组内发往同一目的端的数据包可以被发送到 同一光设备的不同物理端口或不同光设备的物理端口。 所述 MP4组内发往同 一目的端的数据包可以由光传输网络中的任一光设备汇聚后再转发给目的端 侧的数据设备或直接转发给目的端侧的数据设备。
图 3为实现本发明实施例多端口负载分担方法的网络模型示意图。如图 3 所示， 数据设备 101 包括 2个逻辑端口和 2个物理端口。 其中， 所述 2个逻 辑端口分别标记为 D-L1和 D-L2 ,所述 2个物理端口分别标记为 D-P1和 D-P2 , 并且每个物理端口分别对应两个物理子通道， 具体为： 物理端口 D-P1对应物 理子通道 SC-1和 SC-2, 物理端口 D-P2对应物理子通道 SC-3和 SC-4。 数据 设备 101与光传输网络中的光设备 102相连， 数据设备 101的物理端口 D-P1 与光设备 102的物理端口 0-P1对应相连形成一条 U I链路， 数据设备 101 的另一物理端口 D-P1与光设备 102的物理端口 0-P2对应相连形成另一条 UNI 链路。其中，物理子通道 SC-1和 SC-2的负载容量总和可以占据 D-P1至 0-P1 间的 UNI链路的所有带宽， 并且物理子通道 SC-1和 SC-2所占据的带宽比例 可根据具体需求进行调整， 例如可以调整为二者相同或不同， 所述调整可以 釆用静态配置或者动态配置的方式。 物理子通道 SC-3和 SC-4与此类似， 不 再赘述。 这里将物理端口 D-P1和 D-P2配置到一个 MP4组内。
根据本发明实施例所述 MP4方法， 经过数据设备 101逻辑端口 D-L1的 数据包将被封装于物理子通道 SC-1 和 SC-3 中并分别通过对应的物理端口 D-P1和 D-P2进行转发， 所述被封装于物理子通道 SC-1和 SC-3中的数据包 分别沿两条 U I链路发送给所述光设备， 光设备将分别从对应的物理端口 0-P1 和 0-P2接收所述数据包， 并根据所述数据包的业务流向将封装于所述 物理子通道 SC-1和 SC-3中的数据包汇聚后从光设备的物理端口 0-P4发送出 去。 同样的， 经过数据设备 101逻辑端口 D-L2的数据包将被封装于物理子通 道 SC-2和 SC-4中并分别通过对应的物理端口 D-P1和 D-P2进行转发， 所述 被封装于物理子通道 SC-2和 SC-4中的数据包分别沿两条 U I链路发送给所 述光设备， 光设备将分别从对应的物理端口 0-P1和 0-P2接收所述数据包， 并根据所述数据包的业务流向将封装于所述物理子通道 SC-2和 SC-4中的数 据包汇聚后从光设备的物理端口 0-P3发送出去。
在逻辑端口 D-L1和 D-L2处获取用于承载转发数据包的物理子通道的方 法可以釆用现有聚合 ( Trunking )或者 ECMP ( Equity Cost Multi-path, 等价 多路径）等技术， 或者釆用其它可行方案。
实际网络应用中， MP4组内物理端口数可以多于图 3中的两个， 可才艮据 相同的原理配置 MP4组。 可选地， 本发明实施例中所述物理端口所对应的子 通道并不限于物理子通道， 也可以为逻辑子通道或者同时包括物理层传输帧 结构的子通道和逻辑子通道。 具体地， 所述物理端口的信号映射格式可以是 所有能够通道化的接口， 例如 POS接口或者 OTN接口， 所有 POS接口或者 OTN接口的等级速率都适用此设计。 标准组织正在制定的 100GE接口， 釆用 lOxlOGE或者 4x25GE, 也可以实现通道化接口。
不同的信号映射端口和传输速率会有不同的 MP4方案，这种 MP4方案可 以釆用人工静态配置的方式， 也可以釆用由设备智能动态配置的方式。 随着 GMPLS ( General Multi-protocol Label Switching, 通用多协议标签交换技术） 协议在网络设备中的应用， 根据 MP4方案， 通过 GMPLS协议来实现光设备 上的动态交叉连接成为可能。在不支持 GMPLS协议的网络节点上，可以根据 MP4方案， 通过人工配置交叉连接， 可以达到相同的效果。 等。 如果所述负载分担组内包括一个以上物理端口， 则各个物理端口的容量 可以相同或者不相同。
根据物理端口容量、 逻辑端口容量和负载分担比例等因素， 可以大致划 分出四种主要的 MP4组配置方案。下面对这些 MP4组配置方案进行详细描述。
方案一、 端口容量相同， 负载平均分担
如图 3所示， 假设数据设备 101与光设备 102连接的 2个物理端口 D-P1 和 D-P2均为 40G OTN接口。 在数据设备上将 D-P1和 D-P2配置成包含 2个 端口的 MP4组，来自逻辑端口 D-L1和 D-L2的负载在 2个物理端口中平均分 配。 40G OTN接口的负载 ODU3通道化后可由 4个负载 ODU2的物理子通道 组成， 每个物理子通道的负载容量为 10G, 分别标记为 #1 0DU2、 #2 0DU2、 #3 ODU2和 #4 ODU2。
表 1描述了由 2个 40G OTN接口组成的 MP4组的端口负载分担通道关 系， 其中 2个物理端口 D-P1和 D-P2对逻辑端口的负载平均分配，各占 50%。
表 1 40G OTN接口的端口负载均衡配置方案
Figure imgf000011_0001
方案二、 端口容量相同， 负载非平均分担
表 2描述了由 2个 40G OTN接口组成的 MP4组的非均衡负载分担通 道关系， 其中物理端口 D-P1和 D-P2对逻辑端口的负载分担按 3： 1比例实施。
40G OTN接口的端口负载非均衡配置方案
Figure imgf000011_0002
方案三、 不同物理端口容量混合的负载分担
表 1和表 2描述了 MP4组内端口容量相同的负载分配方案， 实际上， 还可以将不同容量的物理端口配置成 MP4组。 表 3列出了 3个 40G OTN接 口与 1个 100G OTN接口的负载分配方案， 100G OTN接口假设可通道化成 10 个 10G, 其中光设备也同样具备 3个 40G端口与 1个 100G端口。
表 3 40G端口与 100G端口混合配置方案
Figure imgf000012_0001
方案四、 逻辑端口与物理端口容量不同的负载分担
当数据设备的物理端口与逻辑端口容量不同时， 仍然可以釆用负载分担 组的方法进行配置。 表 4列出了数据设备有 4个 10G逻辑端口和 2个 40G物 理端口情况下的负载分配方案， 其中光设备有 2个 40G物理端口。
表 4 40G物理端口与 10G逻辑端口的负载配置方案
数据设备 数据设备 光设备
物理端口负载标记
物理端口 逻辑端口 物理端口
D-L1 Ul 0DU2 10G
D-Pl 0DU3 D-L2 U2 0DU2 10G 0- PI 0DU3
40G D-L3 m 0DU2 10G 40G
D-L4 M 0DU2 10G
D-L1 Ul 0DU2 10G
D-P2 0DU3 D-L2 U2 0DU2 10G 0-P2 0DU3
40G D-L3 m 0DU2 10G 40G
D-L4 M 0DU2 10G OTN接口的其它速率， 例如 10G、 100G等， 可以釆用同样的方法配置相 对应的物理子通道； 同理， 对于通道化的 POS接口， 也可以 居相应的原理 配置， 例如 40G的 POS接口， 即 STM-256/OC-768等级， 可以在物理子通道 中以 10G的速率即 VC-4-64c的容量进行负载分担。 在更低速率等级的 POS 接口中， 可以通过物理子通道 VC-4配置负载分担方案。
对于未来可能出现的接口， 只要能够进行物理子通道划分， 不管速率等 级多少， 即可以按本发明的实施例中的方法实施负载分担配置。
在网络应用中， 物理端口组通常包含 2、 4、 8或者 16个端口等， 实际上 根据本发明中的方法， MP4组内的物理端口数可以为大于等于 1的任何数。
在网络应用中， 选择不同卡板内的端口组成 MP4组可以进一步提高链路 的可靠性。
以下主要以跨越光传输网络进行数据传输的网络模型为例， 通过本发明 的实施例多端口负载分担方法在不同场景下的应用对本发明的具体实施方式 做进一步说明。 在以下六个应用实例中， RA、 RB、 RZ和 RX是数据设备， OA、 OB、 OC、 OD、 OE、 OZ、 OX 、 OA1 和 OA2分别是与对应数据设备 相连的光设备。 Al、 A2, Bl、 B2, 以及 Zl、 Z2是对应位于不同位置的 UNI 链路。 各台光设备之间有网络光链路互连。
应用实例一
本应用实例为通过本发明实施例的多端口负载分担方法增强网络可靠性 的一个具体应用。
如图 4 所示， 为未釆用本发明实施例的多端口负载分担方法的恢复路由 示意图。 当数据设备 RA与光设备 OA之间的 U I链路 A1和 A2正常时， 从 数据设备 RA至数据设备 RB的路由为 RA->A1 ->OA->OB->B 1 ->RB。 当 UNI 链路 A1发生故障时， 数据设备 RA至数据设备 RB的路由中断。 由于未使用 本发明 MP4方案， 数据设备 RA感知到上述故障后， 为恢复 RA至 RB之间 的路由， 需要在全网重新进行路由查找计算， 选择的备用路由会经过多个中 间设备，该备用路由是 RA -〉 Α2->ΟΑ->ΟΖ -〉 Z2->RZ -〉 Zl ->ΟΖ->ΟΒ -〉 B2->RB。 如图 5所示， 为釆用本发明实施例的多端口负载分担方法的恢复路由示 意图。 当数据设备 RA与光设备 OA之间的 U I链路 A1和 A2正常时， 由于 釆用了本发明 MP4 方案， 因此从数据设备 RA至数据设备 RB 的路由包括 RA->A1 ->OA->OB->B 1 ->RB和 RA->A2->OA->OB->Bl->RB。 当 UNI链路 Al发生故障时， RA至 RB的路由不会中断。数据设备 RA将根据所述故障更 新本地逻辑端口表中的负载分担组信息。 当 RA接收到发往 RB的数据包时， RA将根据所述数据包携带的 RB的地址信息查找地址端口映射表获得对应逻 辑出端口信息， 然后根据所述逻辑出端口信息查找逻辑端口表获取负载分担 组信息， 此时负载分担组仅包括与 U I链路 A2对应的物理端口， 因此数据 设备 RA将根据预设的负载分担策略将所述数据包封装于所述物理端口对应 的至少一个物理子通道中并通过所述物理端口以及所述 U I链路 A2转发至 所述光设备 OA,再由 OA沿着 OA->OB->Bl->RB的路由将数据包转发给 RB。 并利用原有的 OA-OB的光路进行路由转发至 RB。 在数据包传输过程中， 数 据设备 RA与光设备 OA都不需要保护倒换， 并且 RA至 RB 的路由经由 RA->A2->OA->OB->B 1 ->RB , 因此也无需占用 OA至 RZ , RZ至 RB的网络 资源。
从上述方案比较可知， 釆用 MP4技术的组网在 U I链路发生故障时， 可 以隔离故障在网络上的传播， 将 U I链路故障的影响限制在单个节点上， 在 网络上保持原有的路由， 从而节省网络资源， 避免网络震荡。 本实施例釆用 的方案充分利用数据设备与光设备的互动优势， 无需设备或链路倒换即可达 到保护目的。
应用实例二
本应用实例为通过本发明实施例的多端口负载分担方法来增强 UNI链路 的保护。
如图 3所示，链路 D-P2至 0-P2发生故障，该故障可能是 D-P2端口故障、 0-P2端口故障或者 D-P2与 0-P2两端口之间的光纤故障所致。尽管对应于该 链路的物理子通道 SC-3和 SC-4失效，但通过光设备的物理端口 0-P3传送的 数据包仍然可以封装于物理子通道 SC-2中经由数据设备 101的物理端口 D-P1 转发至光设备的物理端口 0-P1 , 再由光设备的物理端口 0-P1传送至物理端 口 0-P3; 同样， 通过光设备的物理端口 0-P4传送的数据包仍然可以封装于 物理子通道 SC-1中经由数据设备 101的物理端口 D-P1转发至光设备的物理 端口 0-P1 , 再由光设备的物理端口 0-P1传送至物理端口 0-P4。
由此可见， 链路发生故障后， 数据设备与光设备都不需要进行任何倒换 操作即可达到保护目的， 且无需额外的保护资源。
如图 6所示为釆用本发明实施例的多端口负载分担方法的 U I链路保护 示意图。 图 6表明了 MP4保护与 Tmnking (聚合 )保护的区别， 从图中可以 看出， 釆用 MP4组的 U I链路可以保护数据设备与光设备之间的所有 U I 链路， 不受限制； 而 Tmnking保护则有很大的局限性， 只能保护通往相同目 的地址的 UNI链路。
釆用 MP4方案， 组内各个端口互为保护， 端口数越多， 相应的保护端口 数也越多。 即使绝大多数 U I链路出现故障， 只要有一条链路可以使用， 则 通过该数据设备转发的数据包都能通达目的地， 且无需倒换恢复。 被保护的 U I链路可以去往不同目的地。
如图 2所示， 本发明实施例多端口负载分担方法能够实现同一个 MP4组 中不同物理端口对应的物理子通道承担的负载的自适应调整。 在某个物理端 担该物理端口所对应的物理子通道上的负载。 由于目的地相同的数据包是通 过对应于不同物理端口的多个物理子通道发送的， 因而， 光设备的相应物理 端口仍然能够接收到数据包， 且在物理子通道容量冗余度足够时能保持与故 障前相同的流量， 因而能够提高网络的可靠性。
应用实例三
本应用实例为釆用本发明实施例的多端口负载分担方法实现多台网络设 备的 UNI链路保护。
如图 7所示， 以数据设备路由器为例。 路由器 RA有两个物理端口， 分别 标记为 D-P1和 D-P2,这两个物理端口分别连接至不同的光设备 OA1和 OA2。
RA需要与路由器 RB和 RC建立两条链路，同样可以将物理端口 D-P1和 D-P2 配置成 MP4端口组。
路由器 RA可以通过标记为 S1和 S2的两条路由将数据包转发给 RB, 其 中路由 S1为 RA->OAl->OB->RB, 路由 S2为 RA->OA2->OB->RB。 具体地， RA根据预设的负载分担策略将转发给 RB的数据包分别封装于物理端口 D-P1 和 D-P2所对应的至少一个物理子通道中， 然后将封装于物理端口 D-P1所对 应的物理子通道中的数据包通过物理端口 D-P1转发至光设备 OA1 ,并将封装 于物理端口 D-P2所对应的物理子通道中的数据包通过物理端口 D-P2转发至 光设备 OA2, 接着分别通过光设备 OA1和 OA2传送到达光设备 OB后， 由 光设备 OB 将来自这两条不同路由的数据包汇聚后再通过同一光设备物理端 口转发给路由器 RB。 由此可知， 即使路由器 RA和光设备 OA1之间的 UNI 链路出现故障而导致路由 S1中断， 由路由器 RA转发给路由器 RB的数据包 仍然可以利用路由 S2来转发， 即从路由器 RA转发至光设备 OA2, 再通过光 设备 OA2发送给路由器 RB,从而达到保护目的。路由器 RA至路由器 RC的 数据链路保护机制与路由器 RA至路由器 RB的数据链路保护机制相同，可以 通过负载分担路径 S3和 S4进行数据包的传送， 这里不再赘述。
从上述分析可以得知， 当本发明实施例的多端口负载分担方法应用于多 个网络设备的 U I链路保护时， 有更强的保护能力。 因为这种应用不仅可以 保护 U I的链路故障， 还可以保护网络设备的节点故障。 如图 7中的光设备 OA1或者 OA2 , 无论哪台光设备出现故障， 这个故障无论是端口故障还是设 备本身瘫痪， 都可以起到保护作用。 实际网络应用中， 还可以将本发明实施 例的多端口负载分担方法应用于多台数据设备与多台光设备之间的 U I链路 保护。
应用实例四 本应用实例为通过本发明实施例的多端口负载分担方法节省物理端口数 量的具体应用。
在核心骨干网上为了提高网络的可靠性， 通常釆用网状网结构， 网络中 各个节点全部互连， 需要^ ^多的大容量端口， 而业务占用总带宽的实际比例 却并不高。 作为与数据设备相连的光设备， 不仅客户侧需要很多端口， 同样 线路侧也需要庞大数量的物理端口， 因而节省端口数量尤其重要。
图 8为未釆用 MP4方案的端口连接示意图，图 9为釆用 MP4方案的端口 连接示意图。 通过图 8与图 9的对比， 可以看出釆用 MP4方案后， 数据设备 与光设备都可以节省大量物理端口。 图 8、 图 9中是包含 5个节点的网状网， 未釆用 MP4方案前， 每个路由器需要 4个物理端口与光设备连接； 釆用 MP4 方案后， 以 RA为例， 如图 9中所示， RA通过将与 OA相连形成 UNI链路 A1的物理端口和与 OA相连形成 U I链路 A2的物理端口设置为一负载分担 组， 使得从 RA分别发送给 RB、 RC、 RD和 RE中任一数据设备的数据包都 可以封装于 U I链路 A1和 A2分别对应的至少 1个物理子通道中并通过所述 U I链路 A1和 A2发送给光设备 OA后再由 OA转发给对应的数据设备。 由 此可知， 每个路由器只需要 2个配置成负载分担组的物理端口与光设备相连 就可以实现分别与其他 4个路由器相连， 这样不仅可以提供较佳的网络可靠 性， 还可以节省物理端口的数量。
在图 9 所示非常简单的五个节点的网状网中， 路由器可节省的端口数共 达十个； 光设备也节省同样多的端口数。 随着网络节点数的增加， 节省的物 理端口数将更多。
本应用实例中釆用本发明实施例的多端口负载分担方法还能达到减少初 期投资， 平滑扩容的目的。
由图 8和图 9的比较， 还可以看出釆用 MP4方案带来的另一价值。 运营 商建网初期业务容量不大， 但是需要建立可靠性很高的网状网。 如果不釆用 MP4 方案， 则需要建网初期就配置大量的卡板以达到互连的目的， 配置的端 口容量大量闲置降低投资收益； 而釆用 MP4方案后， 只要配置一个大容量端 口就可以达到目的， 当业务容量增加到超过一个端口的容量时再新增一个端 口， 大大提高投资收益率， 真正实现随业务增长的平滑扩容。
应用实例五
本应用实例为釆用本发明实施例的多端口负载分担方法实现数据设备的 互连带宽与端口容量无直接关联的具体应用。
如图 10所示， 路由器 RA只有 10G速率的物理接口， 路由器 RB有 40G 速率的物理接口， 通过 MP4方案后， RA与 RB之间就可以有 40G的带宽互 连。 RB不再需要配置 10G速率的业务卡板。
在目前网络中，数据设备之间的逻辑互连，以及数据设备与光设备的互连， 都要求物理端口容量一致， 才能实现正常连接。 釆用 MP4方案后， 数据设备 之间的逻辑互连带宽可以多种灵活方式连接， 甚至通过物理子通道的会聚以 实现低端产品 （没有大容量端口 )与高端产品以高速率连接。
应用实例六
本应用实例为釆用本发明实施例的多端口负载分担方法实现比 TE ( Traffic Engineering, 流量工程） 更具优越性的隧道方案的具体方式。
在当前 MPLS ( Multi-Protocol Label Switching,多协议标记交换）网络中， 为建立端到端的 TE,需要在路由器上逐跳部署。中间经过路由器的数目越多， 实施的 TE就越复杂， 浪费的资源也越多。
图 11和图 12是釆用 MP4方案前后的对比。图 11、图 12中需要建立 RA-RZ 的遂道， 其中， 图 11 中的路径 S1表示建立遂道所经过的实际物理路径， S2 和 S3表示与隧道共用物理通道但不同子通道的单跳路由。 釆用本实施例技术 方案 MP4后， 建立的遂道可以直接通过光设备的物理子通道 S4实现， 不需 要中间路由器的参与， 节省大量的数据设备资源。
可见， 釆用 MP4 方案后， 不仅节省了大量物理资源， 而且由于通过光层 建立直接通路， 遂道的 QOS也大为提升， 不再受中间路由器的影响。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流 程， 是可以通过计算机程序来指令相关的硬件来完成， 所述的程序可存储于 一计算机可读取存储介质中， 该程序在执行时， 可包括如上述各方法的实施 例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体（ Read-Only Memory, ROM )或随机存 己忆体 ( Random Access Memory, RAM )等。 请参阅图 13 , 本发明的实施例还提供一种用于实现多端口负载分担的数 据设备， 包括：
接收模块 131 , 用于接收发送给目的端的数据包；
地址端口映射表查找模块 132,用于根据所述数据包携带的目的端的地址 信息查找地址端口映射表获得对应逻辑出端口信息；
逻辑端口表查找模块 133 ,用于根据所述逻辑出端口信息查找逻辑端口表 获取对应负载分担组的信息， 其中所述负载分担组包括至少两个物理端口， 每个物理端口对应至少两个物理子通道；
发送模块 134,用于根据预设负载分担策略将所述数据包分别封装于所述 负载分担组内每个物理端口所对应的至少一个物理子通道中并通过对应的物 理端口向目的端转发。
可选地， 本发明实施例中所述物理端口所对应的子通道并不限于物理子 子通道。 具体地， 所述子通道包括 OTN接口对应的 ODU系列、 POS接口对 应的 VC-4系列、 或 100GE接口对应的通道化接口。
本发明实施例所提供的数据设备可以为路由器或交换机等。 该数据设备 可以参照本发明方法实施例中图 2所示的方法实现多端口负载分担。
再请参阅图 14, 本发明的实施例还提供一种用于实现多端口负载分担的 网络系统， 包括数据设备 141和网络设备 142, 其中：
所述数据设备 141 , 用于接收发送给目的端的数据包， 根据所述数据包中 携带的目的端的地址信息查找地址端口映射表获得对应逻辑出端口信息， 并 根据所述逻辑出端口信息查找逻辑端口表获取对应负载分担组的信息， 其中 所述负载分担组包括至少两个物理端口， 每个物理端口对应至少两个物理子 通道， 根据预设负载分担策略将所述数据包分别封装于所述负载分担组内每 个物理端口所对应的至少一个物理子通道中并通过对应的物理端口向网络设 备 142转发； 给目的端或直接发送给目的端。
可选地， 本发明实施例中所述物理端口所对应的子通道并不限于物理子 子通道。
进一步地， 所述网络设备 142 为光设备， 所述数据设备可以为路由器、 交换机等。 本发明的实施例网络系统可以参照本发明方法实施例中图 2 所示 的方法实现多端口负载分担， 可以参照上述应用实例一至应用实例六所示的 组网方式组成各种网络形式。
本发明的实施例提供的数据设备和网络系统， 在同一个 MP4组内， 将负 载分担到一个以上物理端口所分别对应的多个物理子通道上， 因而负载能够 通过不同的子通道到达相同的网络设备端口， 在 U I链路发生故障时， 数据 包能够通过其它的子通道到达目的地址， 因而本发明的实施例提高了网络的 可靠性， 实现连接不同目的地的 U I链路的保护。 由于数据包可以通过多个 物理端口中的多个物理子通道进行传输， 因而本发明的实施例还解决了核心 网大容量端口不足的问题， 在建网时釆用本发明的实施例能够减少初期投资， 还能够实现平滑扩容。
以上所述， 仅为本发明的具体实施方式， 但本发明的保护范围并不局限 于此， 任何熟悉本技术领域的技术人员在本发明揭露的技术范围内， 可轻易 想到的变化或替换， 都应涵盖在本发明的保护范围之内。 因此， 本发明的保 护范围应以权利要求的保护范围为准。

Claims

权 利 要求 书
1、 一种多端口负载分担方法， 其特征在于， 包括：
接收发送给目的端的数据包；
根据所述数据包携带的目的端的地址信息查找地址端口映射表获得对应逻 辑出端口信息；
根据所述逻辑出端口信息查找逻辑端口表获取对应负载分担组的信息， 其 中所述负载分担组包括至少两个物理端口， 每个物理端口对应至少两个子通道； 根据预设的负载分担策略将所述数据包分别封装于所述负载分担组内每个 物理端口所对应的至少一个子通道中并通过对应的物理端口向目的端转发。
2、 根据权利要求 1 所述的多端口负载分担方法， 其特征在于， 所述通过 对应的物理端口向目的端转发的步骤具体包括： 将所述发送给目的端的数据包 通过对应的物理端口发送给传输网络， 并通过所述传输网络根据所述数据包的
3、 根据权利要求 1所述的多端口负载分担方法， 其特征在于， 所述不同物 理端口内的子通道的容量相等或者不相等。
4、 根据权利要求 1所述的多端口负载分担方法， 其特征在于， 各个物理端 口的容量相同或者不同。
5、 根据权利要求 1所述的多端口负载分担方法， 其特征在于， 所述逻辑出 端口的容量与物理端口的容量相同或者不同。
6、 根据权利要求 1所述的多端口负载分担方法， 其特征在于， 所述目的端 的地址信息为目的端的 IP地址， 所述地址端口映射表对应为路由表； 或
所述目的端的地址信息为目的端的 MAC地址， 所述地址端口映射表对应为 MAC表。
7、 根据权利要求 1至 6中任意一项所述的多端口负载分担方法， 其特征在 于， 所述子通道为物理子通道。
8、 根据权利要求 1所述的多端口负载分担方法， 其特征在于， 所述子通道 包括物理层传输帧结构的子通道和逻辑子通道。
9、 一种用于实现多端口负载分担的数据设备， 其特征在于， 包括： 接收模块， 用于接收发送给目的端的数据包；
地址端口映射表查找模块， 用于根据所述数据包携带的目的端的地址信息 查找地址端口映射表获得对应逻辑出端口信息；
逻辑端口表查找模块， 用于根据所述逻辑出端口信息查找逻辑端口表获取 对应负载分担组的信息， 其中所述负载分担组包括至少两个物理端口， 每个物 理端口对应至少两个子通道； 以及
发送模块， 用于根据预设负载分担策略将所述数据包分别封装于所述负载 分担组内每个物理端口所对应的至少一个子通道中并通过对应的物理端口向目 的端转发。
10、 根据权利要求 9所述的数据设备， 其特征在于， 所述子通道包括物理 层传输帧结构的子通道和逻辑子通道。
11、 根据权利要求 9所述的数据设备， 其特征在于， 所述子通道为物理子 通道。
12、 一种网络系统， 其特征在于， 包括数据设备和网络设备， 其中： 所述数据设备用于接收发送给目的端的数据包， 根据所述数据包中携带的 目的端的地址信息查找地址端口映射表获得对应逻辑出端口信息， 并根据所述 逻辑出端口信息查找逻辑端口表获取对应负载分担组的信息， 其中所述负载分 担组包括至少两个物理端口， 每个物理端口对应至少两个子通道， 根据预设负 载分担策略将所述数据包分别封装于所述负载分担组内每个物理端口所对应的 至少一个子通道中并通过对应的物理端口向网络设备转发； 端或直接发送给目的端。
13、 根据权利要求 12所述的网络系统， 其特征在于， 所述子通道为物理子 通道。
14、 根据权利要求 12所述的网络系统， 其特征在于， 所述网络设备为光设 备， 所述数据设备为路由器或交换机。
PCT/CN2010/070632 2009-02-13 2010-02-11 一种多端口负载分担方法、装置和网络系统 Ceased WO2010091640A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP10740929.4A EP2391085B1 (en) 2009-02-13 2010-02-11 Multiple ports load sharing method, apparatus and network system
US13/209,043 US8848717B2 (en) 2009-02-13 2011-08-12 Method, apparatus, and network system for multi-port load sharing

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2009100090261A CN101605091B (zh) 2009-02-13 2009-02-13 一种多端口负载分担方法、装置和网络系统
CN200910009026.1 2009-02-13

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US13/209,043 Continuation US8848717B2 (en) 2009-02-13 2011-08-12 Method, apparatus, and network system for multi-port load sharing

Publications (1)

Publication Number Publication Date
WO2010091640A1 true WO2010091640A1 (zh) 2010-08-19

Family

ID=41470648

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/070632 Ceased WO2010091640A1 (zh) 2009-02-13 2010-02-11 一种多端口负载分担方法、装置和网络系统

Country Status (4)

Country Link
US (1) US8848717B2 (zh)
EP (1) EP2391085B1 (zh)
CN (1) CN101605091B (zh)
WO (1) WO2010091640A1 (zh)

Families Citing this family (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101605091B (zh) * 2009-02-13 2012-02-22 华为技术有限公司 一种多端口负载分担方法、装置和网络系统
CN101854567B (zh) * 2010-05-14 2013-06-12 中兴通讯股份有限公司 光传输层的数据传输方法及装置
CN102065022B (zh) * 2011-02-17 2012-07-25 福建星网锐捷网络有限公司 一种聚合口流量平衡的方法、装置、聚合口及网络设备
CN102082736A (zh) * 2011-03-08 2011-06-01 杭州华三通信技术有限公司 实现非均衡负载分担的方法及装置
CN102209040B (zh) * 2011-07-12 2013-12-25 杭州华三通信技术有限公司 基于多网络端口实现负载分担的方法和装置
CN102647346A (zh) * 2012-03-30 2012-08-22 汉柏科技有限公司 一种对报文分流的方法及系统
CN102647348A (zh) * 2012-03-30 2012-08-22 汉柏科技有限公司 负载分担的方法及系统
CN102647345A (zh) * 2012-03-30 2012-08-22 汉柏科技有限公司 对ipsec数据报文进行负载分担的方法及系统
JP5937424B2 (ja) * 2012-05-29 2016-06-22 株式会社日立製作所 通信装置及びそれを制御する方法
CN102724131B (zh) * 2012-06-29 2016-08-03 杭州华三通信技术有限公司 一种ip报文等价路由均衡实现的方法及装置
CN103780511B (zh) 2012-10-19 2018-09-28 中兴通讯股份有限公司 链路聚合组中流量路径的协商方法及装置
US9225600B2 (en) 2013-01-20 2015-12-29 Lenovo Enterprise Solutions (Singapore) Pte. Ltd. Automatic configuration of host networking device networking interface without user interaction
US9154408B2 (en) * 2013-02-26 2015-10-06 Dell Products L.P. System and method for traffic polarization during failures
US9660836B2 (en) 2014-05-06 2017-05-23 Lattice Semiconductor Corporation Network topology discovery
US9590825B2 (en) 2014-05-09 2017-03-07 Lattice Semiconductor Corporation Stream creation with limited topology information
WO2015172293A1 (zh) * 2014-05-12 2015-11-19 华为技术有限公司 数据传输方法、装置及交换机
US10536389B1 (en) * 2014-11-03 2020-01-14 Amazon Technologies, Inc. Biased selection of dedicated physical connections to provider network
CN105721234B (zh) * 2014-12-05 2019-04-05 杭州迪普科技股份有限公司 端口聚合方法及装置
CN105763472B (zh) * 2014-12-18 2020-08-11 中兴通讯股份有限公司 一种跨板转发的方法和装置
CN106330698B (zh) * 2015-06-24 2020-04-28 中兴通讯股份有限公司 一种局部路由的恢复方法及装置
CN105262696B (zh) * 2015-09-01 2019-02-12 上海华为技术有限公司 一种多路径分流的方法及相关设备
CN107301141A (zh) * 2016-04-14 2017-10-27 大唐移动通信设备有限公司 一种RapidIo系统链路聚合的方法及装置
CN108156065A (zh) * 2017-12-25 2018-06-12 上海京颐科技股份有限公司 基于网桥的医患系统
CN107948045A (zh) * 2017-12-25 2018-04-20 上海京颐科技股份有限公司 基于网桥的医患系统
CN112087376B (zh) * 2019-06-14 2023-03-14 中兴通讯股份有限公司 一种负载分担的方法及装置
CN113285878B (zh) * 2020-02-20 2022-08-26 华为技术有限公司 负载分担的方法、第一网络设备
US20220116192A1 (en) * 2020-10-14 2022-04-14 Newracom, Inc. Multi-link operation and indication for next generation wireless local area networks
CN113709058A (zh) * 2021-07-26 2021-11-26 普联技术有限公司 一种数据传输方法
CN115665047B (zh) * 2022-10-27 2025-07-11 北京火山引擎科技有限公司 报文传输控制方法、装置、介质及设备

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1369991A (zh) * 2002-04-01 2002-09-18 北京港湾网络有限公司 交换机捆绑端口分流方法及交换机芯片
US20070217345A1 (en) * 2006-03-17 2007-09-20 Fujitsu Limited Ring type network system
CN101179510A (zh) * 2007-12-13 2008-05-14 中兴通讯股份有限公司 用于虚拟交换系统的主备链路负荷分担方法和装置
CN101257447A (zh) * 2008-03-14 2008-09-03 华为技术有限公司 一种负载分担的方法、系统及路由装置
CN101605091A (zh) * 2009-02-13 2009-12-16 华为技术有限公司 一种多端口负载分担方法、装置和网络系统

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5959968A (en) * 1997-07-30 1999-09-28 Cisco Systems, Inc. Port aggregation protocol
US6363077B1 (en) * 1998-02-13 2002-03-26 Broadcom Corporation Load balancing in link aggregation and trunking
US6473424B1 (en) * 1998-12-02 2002-10-29 Cisco Technology, Inc. Port aggregation load balancing
US6731599B1 (en) * 1999-07-01 2004-05-04 Nortel Networks Limited Automatic load sharing-trunking
US20020109879A1 (en) * 2000-08-23 2002-08-15 Wing So John Ling Co-channel modulation
US7209436B1 (en) * 2000-12-30 2007-04-24 Redback Networks Inc. Method and apparatus for variable rate pipes
JP3868815B2 (ja) * 2002-01-10 2007-01-17 富士通株式会社 通信システム
US7346277B2 (en) 2003-09-08 2008-03-18 Lucent Technologies Inc. Joint-layer restoration in packet-over-optical networks
US7729361B2 (en) * 2006-02-24 2010-06-01 Cisco Technology, Inc. Method and system for power-efficient adaptive link aggregation
US20070214282A1 (en) 2006-03-13 2007-09-13 Microsoft Corporation Load balancing via rotation of cluster identity
JP5086585B2 (ja) * 2006-08-11 2012-11-28 アラクサラネットワークス株式会社 ネットワーク中継装置
CN101247351B (zh) * 2008-03-18 2011-04-20 杭州华三通信技术有限公司 负载分担方法及装置
CN101355519B (zh) * 2008-09-12 2010-11-10 杭州华三通信技术有限公司 一种实现业务流负载分担的方法和装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1369991A (zh) * 2002-04-01 2002-09-18 北京港湾网络有限公司 交换机捆绑端口分流方法及交换机芯片
US20070217345A1 (en) * 2006-03-17 2007-09-20 Fujitsu Limited Ring type network system
CN101179510A (zh) * 2007-12-13 2008-05-14 中兴通讯股份有限公司 用于虚拟交换系统的主备链路负荷分担方法和装置
CN101257447A (zh) * 2008-03-14 2008-09-03 华为技术有限公司 一种负载分担的方法、系统及路由装置
CN101605091A (zh) * 2009-02-13 2009-12-16 华为技术有限公司 一种多端口负载分担方法、装置和网络系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2391085A4 *

Also Published As

Publication number Publication date
CN101605091A (zh) 2009-12-16
EP2391085B1 (en) 2016-09-28
EP2391085A4 (en) 2011-11-30
US8848717B2 (en) 2014-09-30
EP2391085A1 (en) 2011-11-30
US20110299396A1 (en) 2011-12-08
CN101605091B (zh) 2012-02-22

Similar Documents

Publication Publication Date Title
CN101605091B (zh) 一种多端口负载分担方法、装置和网络系统
US7872991B2 (en) Methods and systems for providing MPLS-based layer-2 virtual private network services
US7751329B2 (en) Providing an abstraction layer in a cluster switch that includes plural switches
US8588060B2 (en) E-spring (G.8032) interworking to provide access protection
CN103312613B (zh) 业务报文转发方法和装置
CN104221336B (zh) 用于虚拟网络架构链路故障恢复的系统和方法
CN1625176B (zh) 基于边缘到边缘伪线仿真协议的通信方法
CN100568787C (zh) 支持以太网媒体接入控制电路的方法及装置
US20070280251A1 (en) Ring Network And A Method For Implementing The Service Thereof
US20210176178A1 (en) Pseudowire protection using a standby pseudowire
US8391180B2 (en) Method to develop hierarchical ring based tree for unicast and/or multicast traffic
US20050105538A1 (en) Switching system with distributed switching fabric
WO2019129236A1 (zh) 基于隧道传输数据的方法及装置
US10673744B2 (en) Methods, systems, and computer readable media for advanced distribution in a link aggregation group
Ali et al. Traffic engineering in metro ethernet
JP2003158539A (ja) ネットワーク転送システム及び転送方法
WO2022063010A1 (zh) 一种业务保护方法和网络节点
CN103188151A (zh) 一种分组传送网络和在分组传送网络中传输业务的方法
US7899061B2 (en) Access device and service transmission method
CN107154896A (zh) 一种数据传输方法以及转发设备
JP4588501B2 (ja) 通信カード及び通信方法
US7466697B1 (en) Link multiplexing mechanism utilizing path oriented forwarding
JP6499624B2 (ja) ネットワーク装置及びフレーム転送方法
JP4883317B2 (ja) 通信システム、ノード、端末、プログラム及び通信方法
Iwata et al. Global open Ethernet architecture for a cost-effective scalable VPN solution

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 10740929

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

REEP Request for entry into the european phase

Ref document number: 2010740929

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2010740929

Country of ref document: EP